Main goal of the proposed research is development of efficient and universal genome manipulation technology based on the tailor-made variants of site-specific recombinase Flp. We aim to (1) identify and classify all DNA sequences in human genome that resemble native recombination target for Flp, FRT;(2) evolve Flp variants that can recombine the corresponding classes of FRT-like sequences;(3) develop an efficient selection system to identify cells in which the desired site-specific recombination took place;and (4) correct genetic disease-causing mutations in model human cell lines using the evolved Flp variants. To identify and classify all FRT-like sites in human genome, we will enhance our existing search program TargetFinder. The modified program will overcome limitations of the current program on the length of DNA to be searched and will be able to screen the longest contigs. After all individual DNA sequences that make up human genome are screened for potential FRT-like sites based on their resemblance to FRT, these sites will be grouped into classes depending on particular sequence patterns of these sites. We will employ structure-based amino acid alterations and directed evolution strategies to evolve Flp variants that can recombine the corresponding classes of FRT-like sequences. All of the evolutionary steps will be carried out by in vitro mutagenesis and gene shuffling, followed by efficient genetic screens in Escherichia coli. Selected representatives of Flp variants will be expressed in bacteria, isolated and analyzed in in vitro recombination reactions. After tests in bacterial system, Flp variants will be assessed in CHO and human cells for mediating integration and gene replacement reactions. To develop efficient selection system for identifying cells in which desired gene replacement reaction took place, we will modify the well-established approach used to identify cells, in which gene replacement has occurred by homologous recombination. Our selection scheme will have two steps. In the first step, we will transfect cells with a construct that contains an antibiotic resistance gene and a suicide gene flanked by the FRT-like sites that correspond to the ones that flank a genomic region to be replaced. This construct will also contain another suicide gene located outside the region flanked by the FRT-like sites. Upon site-specific recombination in some cells, the genomic region of interest will be replaced with the antibiotic/suicide genes cassette. In the second step, the cells selected for antibiotic resistance will be transfected with a construct that contains a DNA fragment for replacement flanked by the corresponding FRT-like sites. The transfected cells will be then treated with prodrugs, which will be converted into toxic products by enzymes coded by the suicide genes. The enzymes will be active in the cells in which second gene replacement did not occur or in which a construct used in the first step integrated into genome randomly by non-homologous recombination. We will experiment with different suicide genes to find their optimal combination that maximizes the yield of cells, in which perfect gene replacement has occurred. We also aim to correct genetic disease-causing mutations in model cell lines that represent (1) sickle cell disease, (2) cystic fibrosis, (3) Tay-Sachs syndrome and (4) Lesch-Nyhan syndrome. To achieve this goal, we will locate FRT-like sites in the vicinity of the disease-causing mutations of the corresponding genes and, using the data obtained under Aim 2, identify Flp variants able to recombine the corresponding FRT-like sites. After testing activity of Flp variants on the corresponding sites in bacterial and mammalian systems, we will accordingly modify the constructs of the gene replacement selection system developed under Aim 3 and replace the disease-causing mutations in the cell lines.